Combining LBP and co-occurrence matrix information to accurate recognition of the optic disc in retinal image

The detection and localization of the optic disc in retinal images is an important step in the diagnosis of ocular diseases and many works with this thematic area were presented over the years. In the current paper we develop a method to accurate detection of the optic disc which combines LBP techniques and features extracted from co-occurrence matrix on color channels. Two classifiers were used to improve the recognition process: a sorter based one, for co-occurrence matrix and a minimum distance based another one, for LBP feature. A number of 100 images from MESSIDOR database were used for testing the proposed algorithm. Comparing with other works the proposed algorithm gave good results in terms of accuracy.